Dear all,
Can OpenSees do a "Direct" Displacement Control? What I mean by that is imposing (cyclic) displacement history directly instead of using load factors to reach a certain displacement as in the displacement-control integrator.
I studied a two-story frame structure tested previously by researchers. They used one actuator for each floor when they imposed a displacement history to the structure. How can I replicate that in OpenSees? I am wondering if I could use the displacement ratio between floors (reported in the paper) to be the load factors for the displacement-control integrator in OpenSees? Most likely, I would get different different results (I guess). So, can OpenSees do a "direct" displacement control similar to that in the above test?
Thanks,
Ronny Purba
"Direct" Displacement Control
Moderators: silvia, selimgunay, Moderators
-
- Posts: 14
- Joined: Tue Dec 07, 2010 8:26 am
- Location: University at Buffalo
- Contact:
-
- Posts: 9
- Joined: Sun May 30, 2010 6:47 am
- Location: Tongji university
Re: "Direct" Displacement Control
1. You can use the "timeSeries" and corresponding "pattern" for each each floor.
For example, Path 1 -dt $dt -filePath displacement.txt -factor 1;
2. Use the SP command.
For example, Path 1 -dt $dt -filePath displacement.txt -factor 1;
2. Use the SP command.
E-mail: 2008wangbin@tongji.edu.cn
-
- Posts: 14
- Joined: Tue Dec 07, 2010 8:26 am
- Location: University at Buffalo
- Contact:
Re: "Direct" Displacement Control
Hi Bin,
Thanks for your prompt response.
I am not quite sure if I could follow your suggestions correctly. For that reason, I have several follow-up questions trying to understand your thoughts.
1. Are you suggesting to try a "very slow" time-history analysis to replicate a cyclic pushover analysis?
2. What type of integrator are you suggesting? I am guessing (since using "timeseries"), one of the transient integrators is suggested. Can we use the static integrator (i.e., displacement control) when using "timeseries"?
3. I read the wiki manual regarding "timeseries", it says "TimeSeries object represents the relationship between the time in the domain, t, and the load factor applied.." Back to my original question "Can I use the displacement ratio between floors to be the load factors here?"
Thanks again Bin. It is greatly appreciated if you could elaborate more your thoughts on using "timeSeries" and corresponding "pattern" for my case.
Ronny Purba
Thanks for your prompt response.
I am not quite sure if I could follow your suggestions correctly. For that reason, I have several follow-up questions trying to understand your thoughts.
1. Are you suggesting to try a "very slow" time-history analysis to replicate a cyclic pushover analysis?
2. What type of integrator are you suggesting? I am guessing (since using "timeseries"), one of the transient integrators is suggested. Can we use the static integrator (i.e., displacement control) when using "timeseries"?
3. I read the wiki manual regarding "timeseries", it says "TimeSeries object represents the relationship between the time in the domain, t, and the load factor applied.." Back to my original question "Can I use the displacement ratio between floors to be the load factors here?"
Thanks again Bin. It is greatly appreciated if you could elaborate more your thoughts on using "timeSeries" and corresponding "pattern" for my case.
Ronny Purba
-
- Posts: 9
- Joined: Sun May 30, 2010 6:47 am
- Location: Tongji university
Re: "Direct" Displacement Control
I give you a example of a three story frame.
---------------------------------------------------------------------------------
timeSeries Path 1 -dt $dt -filePath displacement_1.txt -factor 1;
timeSeries Path 2 -dt $dt -filePath displacement_2.txt -factor 1;
timeSeries Path 3 -dt $dt -filePath displacement_3.txt -factor 1;
pattern Plain 1 1 {
sp 1 1 1
}
pattern Plain 2 2 {
sp 2 1 1
}
pattern Plain 3 3 {
sp 3 1 1}
----------------------------------------------------------------------------------
1. We put the measured displacement value in the files " displacement_*.txt" for each floor. In my example, there are 3 floors in the model.
2. We use constraints Penalty and integrator LoadControl.
3. I find if you use the "load" command, the load ratio can be used. But when we use the "sp" command, the displacement ratio may not be used. We have no choice but to put them in different displacement file for each floor. You can try it.
Bin WANG
---------------------------------------------------------------------------------
timeSeries Path 1 -dt $dt -filePath displacement_1.txt -factor 1;
timeSeries Path 2 -dt $dt -filePath displacement_2.txt -factor 1;
timeSeries Path 3 -dt $dt -filePath displacement_3.txt -factor 1;
pattern Plain 1 1 {
sp 1 1 1
}
pattern Plain 2 2 {
sp 2 1 1
}
pattern Plain 3 3 {
sp 3 1 1}
----------------------------------------------------------------------------------
1. We put the measured displacement value in the files " displacement_*.txt" for each floor. In my example, there are 3 floors in the model.
2. We use constraints Penalty and integrator LoadControl.
3. I find if you use the "load" command, the load ratio can be used. But when we use the "sp" command, the displacement ratio may not be used. We have no choice but to put them in different displacement file for each floor. You can try it.
Bin WANG
E-mail: 2008wangbin@tongji.edu.cn
-
- Posts: 14
- Joined: Tue Dec 07, 2010 8:26 am
- Location: University at Buffalo
- Contact:
Re: "Direct" Displacement Control
Hi Bin,
Thanks for your response.
From your suggestions, I think I have some ideas to try.
I will let you know when I got some results.
Talk to you later. Thanks again.
Ronny Purba
Thanks for your response.
From your suggestions, I think I have some ideas to try.
I will let you know when I got some results.
Talk to you later. Thanks again.
Ronny Purba